Пресс-релиз популярных книг
.
Авторы: 111 А Б В Г Д Е Ж З И Й К Л М Н О П Р С Т У Ф Х Ц Ч Ш Щ Э Ю Я
Книги: 164 А Б В Г Д Е Ж З И Й К Л М Н О П Р С Т У Ф Х Ц Ч Ш Щ Э Ю Я
На сайте 111 авторов, 92 книг, 72 статей, 5913 глав.
30.2 Seismic Random-Excitation Fields
30.2.1 Power Spectral Density of Spatially Varying Ground Acceleration
For multiple excitation problems, the PSD matrices of ground acceleration excitations have the form
Sxx ðivÞ ¼
SX€ 1
X€ 1 ðivÞ SX€ 1
X€ 2 ðivÞ · · · SX€ 1
X€ N ðivÞ
SX€ 2
X€ 1 ðivÞ SX€ 2
X€ 2 ðivÞ · · · SX€ 2
X€ N ðivÞ
· · · · ·· ·· · · · ·
SX€ N
X€ 1 ðivÞ SX€N
X€ 2 ðivÞ · · · SX€N
X€ N ðivÞ
2
6666664
3
7777775
ð30:50Þ
in which
SX€ k
X€ l ðivÞ ¼ rklðivÞ
ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
SX€ k ðvÞSX€ l ðvÞ
q
ð30:51Þ
rklðivÞ ¼ lrklðivÞlexp½iuklðvÞ ð30:52Þ
and rklðivÞ is the acceleration coherence function between the kth and lth supports. Its norm must satisfy
the relation lrklðivÞl # 1: The values of SX€ k ðvÞ and SX€ l ðvÞ in Equation 30.51 can be different due to local
effects. However, earthquake records show that the structural responses due to the difference between
SX€ k ðvÞ and SX€ l ðvÞ are in general of relatively low significance. The factor exp½iuklðvÞ expresses the wave
passage effect, which can be further expressed as
exp½iuklðvÞ ¼ exp½2ivdL
kl=vapp ð30:53Þ
in which dkl is the horizontal distance between the two supports; dL
kl is the projection of dkl along the
propagation direction of the seismic waves; and vapp is the apparent velocity of seismic waves along the
surface. Assume that the time lags between the supports and the origin of the reference coordinate system
are t1; t2; …; tN ; respectively. Without losing generality, let tl $ tk (when l . k). Then dL
kl=vapp ¼ tl 2 tk;
and so Equation 30.53 becomes
exp½iuklðvÞ ¼ exp½ivðtk 2 tlÞ ð30:54Þ
The factor lrklðivÞl reflects the incoherence effect (Kiureghian and Neuenhofer, 1992). Some
mathematical models of rklðivÞ have been established based on practical earthquake records, which will
be shown in the next section. Using Equation 30.51 to Equation 30.54, Equation 30.50 can be written as
SðivÞ ¼ BpJRJB ð30:55Þ
in which
B ¼ diag½expð2ivt1Þ; expð2ivt2Þ; …; expð2ivtN Þ ð30:56Þ
J ¼ diag
ffiffiffiffi
SX€ 1
q
;
ffiffiffiffi
SX€ 2
p
; …;
ffiffiffiffiffi
SX€ N
h q i
ð30:57Þ
Seismic Random Vibration of Long-Span Structures 30-11
© 2005 by Taylor & Francis Group, LLC
R ¼
1 lr12l · · · lr1N l
lr21l 1 · · · lr2N l
· ·· ·· · · · · · · ·
lrN1l lrN2l · · · 1
2
6666664
3
7777775
ð30:58Þ
30.2.2 Several Coherence Models
A number of coherence models have been established based on practical earthquake records. Some of
them are outlined below.
30.2.2.1 Feng – Hu Model
lrklðv; dklÞl ¼ exp½2ðr1v þ r2Þdkl ð30:59Þ
in which r1 and r2 are the coherence parameters. According to the Hai-Cheng earthquake
records (China) and the Niigata earthquake records (Japan), the values of these parameters are
(Feng and Hu, 1981):
Hai-Cheng: r1 ¼ 2 £ 1025 sec/m, r2 ¼ 88 £ 1024 m21
Niigata: r1 ¼ 4 £ 1024 sec/m, r2 ¼ 19 £ 1024 m21
30.2.2.2 Harichandran – Vanmarcke Model
lrklðv; dklÞl ¼ A exp 2
2d
auðvÞ ð1 2 A þ aAÞ
þ ð1 2 AÞexp 2
2d
auðvÞ ð1 2 A þ aAÞ
ð30:60Þ
in which
uðvÞ ¼ K
1 þ ðv=v0Þb21=2 ð30:61Þ
According to the acceleration records of the SMART-1 array (Harichandran and Vanmarcke, 1986), the
parameters in Equations 30.60 and Equations 30.61 are: A ¼ 0:736; a ¼ 0:147; K ¼ 5210;
v0 ¼ 6:85 rad/sec, b ¼ 2:78:
30.2.2.3 Loh – Yeh Model
lrklðv; dklÞl ¼ exp 2a
vdkl
2pvapp
" #
ð30:62Þ
in which a is the wave-number of the seismic waves. According to the 40 acceleration records of the
SMART-1 array (Loh and Yeh, 1988), a ¼ 0:125 is proposed. Parameters vapp and dkl have been explained
above.
30.2.2.4 Oliveira – Hao – Penzien Model
lrklðv; dklÞl ¼ expð2b1dL
kl 2 b2dT
klÞexpð2
a1
ffiffiffiffi
dL
kl
q
2 a2
ffiffiffiffi
dT
kl
q
ðv=2pÞ2Þ ð30:63Þ
in which ai ¼ ð2pai=vÞ þ ðvbi=2pÞ þ ci: dL
kl and dT
kl are the projections of dkl along the propagation
direction of the seismic waves and along its normal direction, respectively. Based on 17 acceleration
records of the SMART-1 array (Oliveira et al., 1991), parameters b1; b2; a1; a2; b1; b2; c1, and c2 are given
for each of these records.
30-12 Vibration and Shock Handbook
© 2005 by Taylor & Francis Group, LLC
30.2.2.5 Luco–Wong Model
lrkll ¼ exp 2
avdkl
vs
2
" #
ð30:64Þ
in which a is the coherence factor, dkl is the horizontal distance between the kth and lth supports and vs is
the shear wave velocity (Luco and Wong, 1986).
30.2.2.6 Qu–Wang–Wang Model
T. Qu, J. Wang, and Q. Wang (QWW) (Qu et al., 1996) proposed the following model
lrkll ¼ exp 2aðvÞdbðvÞ kl
h i
ð30:65Þ
in which
aðvÞ ¼ a1v2 þ a2; bðvÞ ¼ b1v þ b2 ð30:66Þ
a1 ¼ 0:00001678; a2 ¼ 0:001219; b1 ¼ 20:0055; b2 ¼ 0:7674: This model was established based on the
statistics of dozens of records from four closely located arrays with SMART-1 as the leading one. This
model has given reasonable results in some applications in China.
30.2.3 Generation of Ground Acceleration Power Spectral Density Curves
from Acceleration Response Spectrum Curves
In order to carry out the seismic random vibration analysis of important long-span bridges, the local
seismic motion PSD (usually the acceleration PSD) must be established by specialists. For bridges of less
importance, however, such seismic acceleration PSD can be derived from the local ARS curve. Two
methods to perform the transformation are given below.
30.2.3.1 Kaul Method
Kaul (1978) proposed an approximate transformation method that consists of the two equations
Sðv0Þ ¼
4jR2
a ðj; v0Þ
pv0r2 ð30:67Þ
r2 ¼ 2 ln 2
p
v0Ts
ln p
21
ð30:68Þ
in which Raðj; v0Þ is the ARS curve of the seismic absolute acceleration when the damping ratio is j
and the natural angular frequency is v0 (usually j takes 0.05); Sðv0Þ is the equivalent PSD curve
corresponding to the given ARS curve; Ts is the seismic duration; and p is the probability of the peaks that
do not cross with the given positive or negative barriers — usually p ¼ 0:85 is assumed.
30.2.3.2 Iteration Scheme
An iteration-based scheme for the transformation has been proposed (Sun and Jiang, 1990), which
produces more accurate equivalent PSD curve than the Kaul method. For a given ARS curve Raðj; v0Þ; in
order to transform it into an equivalent PSD curve SðvÞ; specify its initial values S0ðviÞ ¼ Si; for
i ¼ 1; 2; …; N: Then the standard deviation of the acceleration response for a SDOF system can be
computed from
s0ðj; v0Þ ¼
ð1
0
SðvÞ
1 þ 4j2ðv=v0Þ2
ð1 2 ðv=v0Þ2Þ2 þ 4j2ðv=v0Þ2 dv
" #1=2
ð30:69Þ
The peak factor r is
r ¼
ffiffiffiffiffiffiffiffiffiffiffi
2 lnðvTsÞ
p
þ 0:5772
. ffiffiffiffiffiffiffiffiffiffiffi
2 lnðvTsÞ
p
ð30:70Þ
Seismic Random Vibration of Long-Span Structures 30-13
© 2005 by Taylor & Francis Group, LLC
in which Ts is the period of the strong earthquake portion. The average zero-crossing rate is
approximately v ¼
ffiffiffiffiffiffiffi
l2=l0 p
p < v0=p:
The acceleration responses computed by means of Equation 30.69 and Equation 30.70 are
Amðj; v0Þ ¼ rs0ðj; v0Þ ð30:71Þ
The percentage errors between Raðj; v0Þ and Amðj; v0Þ can be computed from
Eðv0Þ ¼
lRaðj; v0Þ 2 Amðj; v0Þl
Raðj; v0Þ £ 100% ð30:72Þ
Compute Eðv0Þ for each frequency. If Eðv0Þ is found to be greater than the given tolerance 1 for at least
one frequency, modify all PSD values according to the following equation
Skþ1ðviÞ ¼ Sk ðviÞR2
a ðj; viÞ
A2
mðj; viÞ; i ¼ 1; 2; …; N ð30:73Þ
and then repeat the computations of Equation 30.69 to Equation 30.72. The above process is continued
until Equation 30.72 is satisfied at all frequencies.
30.2.4 Seismic Equations of Motion of Long-Span Structures
For long-span structures subjected to differential ground motion, the equations of motion in the global
coordinate system (assumed to be fixed to the center of the Earth) can be written in partitioned form as
Ms Msb
MT
sb Mb
" #
x€s
x€b
( )
þ
Cs Csb
CT
sb Cb
" #
x_s
x_b
( )
þ
Ks Ksb
KT
sb Kb
" #
xs
xb
( )
¼
0
pb
( )
ð30:74Þ
in which the subscript m represents the master DoF, that is, the support displacements, while the
subscript s represents the slave DoF. The absolute displacement vector xs can be decomposed into the
two parts
xs ¼ ys þ yr ð30:75Þ
where ys is the quasi-static displacement vector (Clough and Penzien, 1993), which satisfies
ys ¼ 2K21
s Ksbxb ð30:76Þ
Substituting Equation 30.75 and Equation 30.76 into Equation 30.74 gives
Msy€r þ Csy_r þ Ksyr ¼ MsK21
s Ksbx€b þ ðCsK21
s Ksb 2 CsbÞx_b ð30:77Þ
It should be pointed out that Equation 30.77 cannot be reduced to the conventional Equation 30.24
when xb represents uniform ground displacements (Clough and Penzien, 1993). This is because Equation
30.74 assumes the damping forces to be proportional to the absolute velocity vector {x_T
s ; x_Tb
}T: In order to
avoid this inconsistency, the damping forces should be assumed to be proportional to the relative velocity
vector {y_T
r ; 0}T in Equation 30.74. This leads to the equations
Msy€r þ Csy_r þ Ksyr ¼ MsK21
s Ksbx€b ð30:78Þ
for uniform ground motion
x€b ¼ ebx€g ð30:79Þ
Note that the following rigid displacement condition is satisfied:
Ks Ksb
KT
sb Kb
" #
es
eb
( )
¼
0
0
( )
ð30:80Þ
30-14 Vibration and Shock Handbook
© 2005 by Taylor & Francis Group, LLC
Its second half gives
Ksbeb ¼ 2Kses ð30:81Þ
Substituting Equation 30.79 into Equation 30.78 and using Equation 30.81 gives Equation 30.24.
30.2.5 Seismic Waves and Their Geometrical Expressions
Seismic waves can be divided into body waves and
surface waves. Body waves include longitudinal
waves (or pressure waves, primary waves or P
waves) and transverse waves (or shear waves,
secondary waves or S waves). Surface waves include
Rayleigh waves and Love waves. For P waves, the
soil particles move parallel to the traveling direction
of waves; for S waves, however, their motion
is normal to the wave traveling direction
(see Figure 30.4). For horizontal shear waves
(SH waves), all particles move horizontally.
For vertical shear waves (SV waves), all particles
move vertically.
Assume that both x and y axes lie in the
horizontal plane. The angle between axis x and the
horizontal traveling direction of these waves is b;
as shown in Figure 30.5. Thus, the acceleration
components along the coordinate axes can be
expressed by the components parallel, or normal,
to the wave traveling direction, that is, for P waves
x€i ¼ u€ i cos b; y€i ¼ u€ i sin b;
z€i ¼ 0 ð30:82Þ
for SH waves
x€i ¼ 2u€ j sin b; y€i ¼ u€ j cos b;
z€i ¼ 0 ð30:83Þ
and for SV waves
x€i ¼ 0; y€i ¼ 0; z€i ¼ u€ k ð30:84Þ
In the equations of motion of an multi-DoF system under uniform ground excitations, that is, Equation
30.24, e is the index vector of inertia forces. Its mathematical expressions for the different waves are:
for P waves
e ¼ ex cos b þ ey sin b ð30:85Þ
for SH waves
e ¼ 2ex sin b þ ey cos b ð30:86Þ
for SV waves
e ¼ ez ð30:87Þ
Clearly, for P waves, e ¼ ex when b ¼ 0 and e ¼ ey when b ¼ 908:
β x
z
y
P
SH
SV
FIGURE 30.4 Particle motion directions for P and S
waves.
x
y
SH
üj
üi
xi
P
b
yi
..
yi
..
..
xi
..
FIGURE 30.5 Transform of ground acceleration
components.
Seismic Random Vibration of Long-Span Structures 30-15
© 2005 by Taylor & Francis Group, LLC
If a structure has N supports, its ground acceleration excitations along the wave traveling direction can
be expressed by the N-dimensional vector
u€ b ¼ {u€ 1; u€ 2;…; u€N}T ð30:88Þ
The m-dimensional ground acceleration vector in Equation 30.78 is
x€b ¼ {x€1; x€2;…; x€m}T ð30:89Þ
The relation between these two vectors is
x€b ¼ EmNu€ b ð30:90Þ
in which EmN is a m £ N block-diagonal matrix
EmN ¼ diag½eb ; eb ; …; eb m£N ð30:91Þ
If only three translations are considered for each support, then m ¼ 3N and each submatrix eb
would be
cos b
sin b
0
8>><
>>:
9>>=
>>;
;
2sin b
cos b
0
8>><
>>:
9>>=
>>;
and
0
0
1
8>><
>>:
9>>=
>>;
for the P, SH and SV waves, respectively.
Using Equation 30.90, Equation 30.76 and Equation 30.78 can be rewritten as
ys ¼ 2K21
s KsbEmN ub ð30:92Þ
Msy€r þ Csy_r þ Ksyr ¼ MsK21
s KsbEmNu€ b ð30:93Þ
Популярные книги
- Старинные занимательные задачи
- Медоносные растения
- Algebratic geometry
- Workbook in Higher Algebra
- Математика Древнего Китая
- Finite element analysis
- Пчеловодство
- Mathematics and art
- Fields and galois theory
- Black Holes
Популярные статьи
- Higher-Order Finite Element Methods
- Электровакуумные приборы
- Riemann zeta functionS
- Универсальная открытая архитектурно-строительная система зданий серии Б1.020.1-71
- Complex Analysis 2002-2003
- Пример расчета прочности елементов, стыков и узлов несущего каркаса здания
- Составы, вещества и материалы для огнезащитыметаллических консрукций и изделий
- CMOS Technology
- Рекомендации по расчету и конструированию сборных железобетонных колонн каркасов зданий серии Б1.020.1-7 с плоскими стыками ВИНСТ
- Советы старого пчеловода